代码拉取完成,页面将自动刷新
同步操作将从 src-openEuler/tensorflow 强制同步,此操作会覆盖自 Fork 仓库以来所做的任何修改,且无法恢复!!!
确定后同步将在后台操作,完成时将刷新页面,请耐心等待。
From 96f364a1ca3009f98980021c4b32be5fdcca33a1 Mon Sep 17 00:00:00 2001
From: Laura Pak <lpak@google.com>
Date: Mon, 2 Aug 2021 13:27:01 -0700
Subject: [PATCH] Validate axis input in tf.raw_ops.QuantizeAndDequantizeV4Grad
PiperOrigin-RevId: 388291385
Change-Id: I3bab68dc61d935afa96c0da021a7b722c6dc8dc8
---
tensorflow/core/kernels/quantize_and_dequantize_op.cc | 7 +++++++
1 file changed, 7 insertions(+)
diff --git a/tensorflow/core/kernels/quantize_and_dequantize_op.cc b/tensorflow/core/kernels/quantize_and_dequantize_op.cc
index 540d900f9f869..d63a49a04be62 100644
--- a/tensorflow/core/kernels/quantize_and_dequantize_op.cc
+++ b/tensorflow/core/kernels/quantize_and_dequantize_op.cc
@@ -158,6 +158,13 @@ class QuantizeAndDequantizeV4GradientOp : public OpKernel {
Tensor* input_backprop = nullptr;
OP_REQUIRES_OK(ctx,
ctx->allocate_output(0, input.shape(), &input_backprop));
+ OP_REQUIRES(
+ ctx, axis_ >= -1,
+ errors::InvalidArgument("Axis must be at least -1. Found ", axis_));
+ OP_REQUIRES(ctx, (axis_ == -1 || axis_ < input.shape().dims()),
+ errors::InvalidArgument(
+ "Axis should be -1 or 0 or a positive value less than ",
+ input.shape().dims(), "but given axis value was ", axis_));
OP_REQUIRES(
ctx, input.IsSameSize(gradient),
此处可能存在不合适展示的内容,页面不予展示。您可通过相关编辑功能自查并修改。
如您确认内容无涉及 不当用语 / 纯广告导流 / 暴力 / 低俗色情 / 侵权 / 盗版 / 虚假 / 无价值内容或违法国家有关法律法规的内容,可点击提交进行申诉,我们将尽快为您处理。